Newest papers

LongZero: Enhancing Long-Context Reasoning in LLMs from Zero Data

Zhennan Wan, Jianfei Chen · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.30132 · cited by 0 works (OpenAlex) · open-access PDF

French

Le renforcement de l’apprentissage à long terme (RL) est un moyen efficace de renforcer le raisonnement à long terme des grands modèles linguistiques (LLM), mais il a besoin de données de formation qui sont longues, variées, difficiles et vérifiables. L’annotation humaine de ces données est coûteuse, les données naturelles sont trop rares pour maintenir le RL, et les méthodes de synthèse existantes dépendent d’un corps externe ou de modèles enseignants plus forts, ce qui devient plus difficile à obtenir à mesure que les modèles formés deviennent plus forts. Plus récemment, le RL à partir de données zéro, dans lequel un modèle apprend des tâches qu’il se propose lui-même, a obtenu des performances excellentes, mais les méth

Machine translation (Meta M2M100); the English original follows.

Abstract

Reinforcement learning (RL) is an effective way to strengthen the long-context reasoning of large language models (LLMs), but it needs training data that are long, diverse, challenging, and verifiable. Human annotation of such data is expensive, naturally occurring data are too scarce to sustain RL, and existing synthesis methods depend on external corpora or stronger teacher models, which become harder to obtain as the trained models get stronger. Recently, RL from zero data, in which a model learns from tasks it proposes itself, has achieved excellent performance, but existing methods target short-context tasks. Therefore, we propose $\textbf{LongZero}$, a method that lets an LLM improve its own long-context reasoning without any external data. Instead of writing questions and answers over existing documents, LongZero first has the model synthesize the question and answer with its supporting evidence in a compact structured form, and only then separately expands the evidence into a long multi-document context. LongZero combines these self-written multi-hop reasoning tasks with self-written multi-round retrieval tasks and trains the model with GRPO. We evaluate LongZero on two LLMs across five long-context reasoning benchmarks and compare it with RL on data synthesized from external corpora. Experimental results show that LongZero can generalize to real tasks, and outperforms both baselines, suggesting a viable and promising path to improving long-context reasoning of LLMs.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Artificial intelligence, Annotation, Path (computing), Reinforcement learning, Zero (linguistics)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange